How bad is it to be uninsured today?

Ezra Klein raised a big stir by suggesting that the possible failure of the health care bill will cost a large number of lives; he cited a figure of 20,000 per year.  (You'll find pushback from Michael Cannon on the number.)

Rather than disputing the number, my question is a simpler one.  Let's say the figure were a correct one.  How would you fill in the following blank?:

"Being uninsured in 2009 is, in terms of life expectancy, as bad as being insured in the earlier year ????"

What is the correct year for this comparison to hold?

Simply knowing the correct year is my main concern in this post, but there is an additional angle.  Twenty years from now there will also be some uninsured Americans, even if the current bill passes.  There will be pleas to help them.  If you wish to help them, does that mean that the insured today also deserve additional health care subsidies?  Or is the whole comparison  just about equality?  How about caring about inequality across time?  If you favor additional subsidies for the uninsured today, are you also committed to wishing there had been additional subsidies for the insured back in year ????

I thank Bryan Caplan for a useful conversation related to this blog post.


Given that a significant degree of the problems regarding unequal/sub-par healthcare relate to it being a positional good, it doesn't seem unreasonable for proponents to focus on the here and now, whatever the may have 'preferred' historically. One position does not necessarily commit you to the other.

The first question I have no idea. However, I dont understand how it is relevant to, well, anything really.

Twenty years from now I believe the life expectancy of an uninsured American i higher than an insured American today. One would hope the technological progress continues. But to compare, say poor people today, with rich people X years ago is strange. As Brad deLong once wrote (from my memory) all middle class Americans are today richer and more powerful than Napoleon. How such a comparison is relevant in any policy debate, I fail to understand. Would be great if you could follow up and enlighten me.

Indeed. If we don't have an uninsured class, then by defintion the uninsured won't have a higher death rate.

This seems like a dandy way to eliminate racial disparity as well. How about we only include "White" on the next census form? Get rid of those pesky underclasses. We're all in this together.

Let's eliminate the term "homeless" too, because I'm pretty sure the number of bedrooms divided by the number of Americans is greater than 1. Mathematically, no need to have an underprivileged category. On average, we all have a place to sleep at night.

If liberals don't give me one of the many billions they are spending, I will die sooner than later. I blame them for that. Now I'll wait for Tyler to ask his readers how soon I'll die and how much humanity is going to lose.

Not sure I understand the question, or its relevance. There are other factors related to life expectancy, such as smoking, obesity, etc. These are often associated with characteristics that are tied to insured/uninsured. For instance, a poor person is more likely to smoke, and far more likely to be uninsured. So part of that person's decreased statistical life span is associated with lack of insurance, but part of it is also tied to the smoking.

So you can't just measure the lifespan of insured people, compare it to the lifespan of uninsured people, and claim that the difference is due entirely to the insurance.

Trying to extand the question across time complicates things even further. Say I am your statistical insured person today. Being from today, I am far less likely to smoke than the 1956 version of me would have been, whether I am insured or not. So part of my increased life expectancy is due to better health care options. But part of it is due to the fact that I don't smoke. On the other hand, I am, far more likely to be obese.

Are we controlling for thee kinds of confounding factors? Should we be?

So Ezra Klein thinks lack of healthcare reform is 25x more deadly than the Iraq War?

I wonder how he rationalizes that belief in his head -- all the years of vile, hateful writing against the latter, but just a few plaintive pleas against the former.

Cognitive dissonance: it ain't pretty.

I'm a 1930 Libertarian.

Jim, how do you figure that Ezra Klein thinks lack of health care reform is "25x more deadly than the Iraq war"? I've seen estimates that number of uninsured in this country is in the area of 30-40 million people. We've had something like 150,000 soldiers in Iraq.

If not having insurance leads to 22,000 more deaths a year, we need to figure out what that incremental death rate was. Ezra's chart shows that in 2006, 18.7% of the 157.7 million 25-64 year olds were uninsured. That's about 29.5 million people. 22,000 incremental deaths means that not having insurance raises the death rate by 74.6 per 100,000 people (22,000 * 29,500,000 * 100,000).

According to the National Vital Statistics Report, the age-adjusted death rate for all Americans fell from 878.1 per 100,000 in 1997 to 776.5 in 2006. In 2001, it was 854.5 or approximately 75 deaths per 100,000 lower – equivalent to the non-insurance/insurance difference. (For contrast, it was ~200 higher in 1980 and over 1000 higher in 1940, so this is a long-term trend.)

Now, of course, these are comparing different populations (Ezra's 24-65 year olds vs. age-adjusted for all ages). The NVSR report doesn't show long-term trends in age banded death rates. But much of the advances in death rates have come from the under 1 year and over 55 crowd. The death rate in the 55-64 crowd has gone down from 1,005 to 891 per 100,000 from 1999 to 2006, while the other insurable age groups have stayed fairly constant during that time. Since 55-64 year olds are about 20% of all 25-64 year olds and the other rates stayed approximately constant, you can estimate the death rate for the insurable crowd dropped about 20 in the last 7 years of data. So I think it's safe to say that you don't have to go back too far to no matter which way you cut it to be equivalent to the insurance differential.

20,000 is not a particularly large number, when you consider that obesity leads to 400,000 deaths per year. Would it be cheaper for the government to ensure everyone, or cheaper for them to reduce the number of people who die from obesity by 5%?

I posted a piece about it here:

Even accepting Ezra Klein's logic, isn't this still a flawed statement? Even if universal insurance would save 20,000 lives, that really tells us nothing about the net result of the health care bill that is being debated. You could agree with Ezra's statement & still be against this obnoxious bill. This seems to be at the heart of much support for the bill - the apparent belief that any bill entitled "Health Care" will magically solve all of our problems with no mitigating side-effects.

It seems Tyler's point is to ask: "What is your goal here--'equality' or attaining some absolute level of well-being?"

If I'm shipwrecked and drowning, and you come by, are you obliged to help me? The answer, I think, is: it depends. If you have a boat and can save me without undue cost to yourself, then yes, you should help me and you're rotten if you don't. But if you're just barely scraping by and helping me would impoverish or imperil you or your loved ones, then the case for your obligation to me is very weak.

I think that's how it works with health care, too. As technology improves, the frequency of situations where we can make a difference without "undue cost" increases, and therefore the quality of care we're expected to provide for each other increases too.

Jim's also implicitly stating that "Deaths we cause don't count if they're not Americans."

Are there uninsured people in Canada? Switzerland. Maybe a comparison of their outcomes?

Won't everyone, in a way, be insured if there are no pre-existing conditions?

Stephen, I realize that this is ancillary to the conversation, but the "400,000 deaths caused by obesity" number is simply false. The CDC has revised it down several times. The currently accepted number is somewhere around 125,000. However, overweight also prevents about 85,000 deaths a year, so the net number of deaths attributable to excess weight is only about 30,000.

However, overweight also prevents about 85,000 deaths a year,

cite. i'm in the healthcare field and I have never heard this before.

As Brad deLong once wrote (from my memory) all middle class Americans are today richer and more powerful than Napoleon. How such a comparison is relevant in any policy debate, I fail to understand.

Obviously it means that we can raise tax rates to 80% and people will STILL be richer than Napoleon.

People who are overweight but not obese have a lower risk of death than those of normal weight, federal researchers are reporting today.

mulp - also keep in mind that other nations have been spending significantly more each year per capita than they did the year before to insure everyone. Convergence is nigh.

Cardiac ablation facility charge: $50,000. Insurance company paid $14800, insured payed $250. Facility was satisfied.

Without insurance, they would have had the meanest, most determined bill collectors after the uninsured for the full $50K.

That's not how it works. You tell the hospital you're uninsured and they'll knock that down to more like $30k. Still sucks, though.

"... we can afford it."

You're kidding, of course.

Arguing against the logic of this proposal is pointless, because there is no logic to it. It’s simple shock doctrine practice. Take a crisis and exploit it to achieve your political goals. They’re goal isn’t to improve the economy. They like the economy just the way it is, because they’re scooping up bargains in real estate right now. All they want is to score a political victory and stick a knife in the poor. This is Hippie Punching 101.

Comments for this post are closed